[#32] Parse 7XX fields as mappings

scriptotek · Jul 6, 2017 · 3cc7175 · 3cc7175
1 parent 16b2531
commit 3cc7175
Show file tree

Hide file tree

Showing 4 changed files with 71 additions and 3 deletions.
diff --git a/README.rst b/README.rst
@@ -260,6 +260,7 @@ MARC21XML                                                    RDF
 ``681`` Subject Example Tracing Note                        ``skos:example``
 ``682`` Deleted Heading Information                         ``skos:changeNote``
 ``688`` Application History Note                            ``skos:historyNote``
+``7XX`` Heading Linking Entries                             ``skos:xxxMatch``  (see below)
 ==========================================================  ===================================================================
 
 Notes:
@@ -268,7 +269,15 @@ Notes:
    classification scheme has been defined in the config.
 
 * SKOS relations are generated from 5XX fields if the fields contain a ``$0``
-  subfield containing either the control number or the URI of the related record.
+  subfield containing either a control number or an URI for the related record.
   The relationship type is ``skos:broader`` if ``$w=g``, ``skos:narrower`` if ``$w=h``,
   and ``skos:related`` otherwise.
-  If  ``$w=r`` and ``$4`` contains an URI, that URI is used as the relationship type.
+  If ``$w=r`` and ``$4`` contains an URI, that URI is used as the relationship type.
+
+*  Mappings/relationships are generated for 7XX headings if the fields contain a ``$0``
+   subfield containing either the control number or the URI of the related record.
+   If ``$0`` contains a control number, an URI pattern for the vocabulary
+   (found in indicator 2 or ``$2``) must be defined in mc2skos.record.CONFIG.
+   If ``$4`` contains an URI, that URI is used as the relationship type.
+   Otherwise, if ``$4`` contains one of the ISO 25964 relations, the corresponding
+   SKOS relation is used. Otherwise, the default value ``skos:closeMatch`` is used.
diff --git a/examples/noubomn-c000011.ttl b/examples/noubomn-c000011.ttl
@@ -11,6 +11,9 @@
     dcterms:identifier "REAL000011" ;
     dcterms:modified "2016-06-23"^^xsd:date ;
     owl:deprecated true ;
+    skos:closeMatch <http://id.loc.gov/authorities/subjects/sh85086566> ;
+    skos:exactMatch <http://data.ub.uio.no/humord/c08221>,
+        <http://www.wikidata.org/entity/Q159341> ;
     skos:inScheme <http://data.ub.uio.no/realfagstermer/> ;
     skos:prefLabel "Mugg"@nb .
 
diff --git a/examples/noubomn-c000011.xml b/examples/noubomn-c000011.xml
@@ -17,4 +17,19 @@
   <datafield ind1=" " ind2=" " tag="150">
     <subfield code="a">Mugg</subfield>
   </datafield>
+  <datafield ind1=" " ind2="7" tag="750">
+    <subfield code="a">Muggsopp</subfield>
+    <subfield code="0">(No-TrBIB)HUME08221</subfield>
+    <subfield code="2">humord</subfield>
+    <subfield code="4">=EQ</subfield>
+  </datafield>
+  <datafield ind1=" " ind2="7" tag="750">
+    <subfield code="a">muggsopp</subfield>
+    <subfield code="0">http://www.wikidata.org/entity/Q159341</subfield>
+    <subfield code="4">http://www.w3.org/2004/02/skos/core#exactMatch</subfield>
+  </datafield>
+  <datafield ind1=" " ind2="0" tag="750">
+    <subfield code="a">Molds (Fungi)</subfield>
+    <subfield code="0">sh85086566</subfield>
+  </datafield>
 </record>
diff --git a/mc2skos/record.py b/mc2skos/record.py
@@ -4,6 +4,7 @@
 from datetime import datetime
 import logging
 from iso639 import languages
+from rdflib import URIRef
 from rdflib.namespace import SKOS
 
 from .constants import Constants
@@ -696,7 +697,7 @@ def parse(self, options):
                     elif sf_w == 'h':
                         relation = SKOS.narrower
                     elif sf_w == 'r' and is_uri(sf_4):
-                        relation = sf_4
+                        relation = URIRef(sf_4)
                     else:
                         relation = SKOS.related
 
@@ -754,3 +755,43 @@ def parse(self, options):
         # madsrdf:historyNote
         for entry in self.record.all('mx:datafield[@tag="688"]'):
             self.historyNote.append(entry.stringify())
+
+        # 7XX: Heading Linking Entries
+        for heading in self.get_terms('7'):
+            sf_4 = heading['node'].text('mx:subfield[@code="4"]')
+            sf_0 = heading['node'].text('mx:subfield[@code="0"]')
+
+            if sf_4 is not None and is_uri(sf_4):
+                relation = URIRef(sf_4)
+            else:
+                relation = {
+                    '=EQ': SKOS.exactMatch,
+                    '~EQ': SKOS.closeMatch,
+                    'BM': SKOS.broadMatch,
+                    'NM': SKOS.narrowMatch,
+                    'RM': SKOS.relatedMatch,
+                }.get(sf_4)
+
+            relation = relation or SKOS.closeMatch  # default
+            if is_uri(sf_0):
+                self.relations.append({
+                    'uri': sf_0,
+                    'relation': relation,
+                })
+            else:
+                scheme_code = {
+                    '0': 'a',  # Library of Congress Subject Headings
+                    '1': 'b',  # LC subject headings for children's literature
+                    '2': 'c',  # Medical Subject Headings
+                    '3': 'd',  # National Agricultural Library subject authority file
+                    '4': 'n',  # Source not specified
+                    '5': 'k',  # Canadian Subject Headings
+                    '6': 'v',  # Répertoire de vedettes-matière
+                    '7': heading['node'].text('mx:subfield[@code="2"]'),  # Source specified in subfield $2
+                }.get(heading['node'].get('ind2'))
+
+                self.append_relation(
+                    ConceptScheme(scheme_code, AuthorityRecord),
+                    relation,
+                    control_number=sf_0
+                )